Campaign leadership / AI systems
I started my career helping campaigns win in some of the country’s most competitive states by finding the facts that could change a race and working with talented people to make sure voters heard them.
Today, I build AI tools for the people doing that work. Campaign staffers are asked to move faster every year, often while fighting with systems that waste their time and wear them down. I want to give some of that time and energy back, so they can focus on the decisions, relationships, and ideas that actually move a campaign.
I know where the time goes because I have done the work myself. I have built research departments, managed teams through difficult races, and helped campaigns respond when the facts, the timeline, or the entire plan changed overnight. The best work has always come from giving good people useful tools, clear expectations, and room to do what they do best.
This map shows the states and districts where I’ve worked. Beneath the data are stories of candidates who were counted out, teams that stayed disciplined, and victories that looked improbable until they happened.
…It was a surprising and swift ascent for the mild-mannered career politician who was previously part of a crop of lower-polling Democratic candidates that party chair Rusty Hicks was publicly pressuring to drop out of the race.
CalMatters
Ruben Gallego outperformed Harris, dominated with Latinos and won Arizona.
NBC News
Final election results: 2023 was the closest Mississippi governor's race since 1999.
Mississippi Today
AZ-01 · AZ-06 · KS-03 · MT-02 · NE-02 · NV-01 · NV-03 · NV-04 · OR-04 · OR-05 · OR-06 · WA-08
No Democratic president with control of the Senate has ever duplicated the achievement of picking up Senate seats, or even holding steady, in a midterm election. Until now?
The Atlantic
GA · IA · KS · MI · NH
Democrats won both Georgia Senate seats and with them, the U.S. Senate majority, serving President Donald Trump a stunning defeat in his last days in office.
AP via PBS NewsHour
Democratic Sen. Joe Manchin has won re-election in West Virginia in the heart of Trump country, dispatching Republican state Attorney General Patrick Morrisey.
Roll Call
Campaign staffers should not have to choose between moving quickly and getting the details right. I build open-source tools that remove repetitive work while keeping the underlying evidence close at hand.
Most of them began with a problem I had encountered myself: legislation scattered across fifty different systems, campaign-finance records that were difficult to search, political ads that could not be reviewed at scale, or decades of public documents trapped in unusable formats.
I share the tools because good infrastructure should not be reserved for the largest campaign or the best-funded firm. If something I built saves another researcher an afternoon, that is an afternoon they can spend thinking, helping their team, or simply going home at a reasonable hour.
Search bills, full text, votes, and legislative history across all 50 states and Congress.
Query FEC data on candidates, contributions, spending, and independent expenditures.
Pull political-ad buys from FCC public inspection files and turn them into structured data.
Astro Teller says "most of why we waste time at work is fear."
Two paintings about obsession, and the labor inside them.
Most people using AI at work are delegating projects they couldn't delegate to a person, then getting frustrated when the computer can't read their mind.
AI is useful when it respects the person using it. I do not believe in replacing political judgment with a black box or adding technology because it sounds impressive. I begin with the work as people actually experience it: where they get stuck, what absolutely has to be right, and what they wish they had more time to do. Then I build a reliable system that gives them some of that time back.
Campaign work is full of surprises, and everything takes longer than anyone expects. The technology matters, but clear priorities, honest expectations, and calm collaboration are what keep people moving when the plan changes.
Outside work
I’m often at the National Gallery, and I still think about visits to the Louvre, the V&A, and Vienna’s Kunsthistorisches Museum. Next up is the Met for Giacometti in the Temple of Dendur. I’m hoping 2027 is the year I make it to the Prado.
I occasionally write about paintings, too. “Staring at Saint Jerome” is one example.
If you want to talk about a campaign, a research problem, or a useful tool, get in touch.
github.com/sh-patterson linkedin.com/in/shawn-patterson